Deep Learning and Artificial Intelligence Engineering Bootcamp in Barcelona Code School

In 13 weeks, go from Python to building LLM-powered AI systems with 1:1 guidance from a Cambridge-trained researcher

Build real ML and AI systems, not just use them

Build RAG pipelines, fine-tuned LLMs, and production ML workflows — not toy examples.
Work with Hugging Face, LangChain, and LlamaIndex — the actual tools used in AI engineering roles.
Learn 1:1 with a Cambridge PhD researcher, not in a cohort of 50 where questions go unanswered.

FORMAT

Remote, mentored

OUTCOME

Junior ML/AI engineering readiness

LANGUAGE

English

DURATION

13 weeks, part-time

Tools

Python, NumPy, Pandas
Scikit-Learn
TensorFlow / Keras
Hugging Face Transformers
LangChain
LlamaIndex
SpaCy
OpenAI API
Google Colab / GCP
Matplotlib, Seaborn

What you'll learn

Building and fine-tuning LLMs with Hugging Face and Mistral
RAG pipelines: embeddings, vector retrieval, and document QA
LangChain and LlamaIndex for AI application development
LLMOps: deploying models to GCP, quantization, and production workflows
Convolutional and recurrent neural networks for vision and text
Classical ML, tree-based models, and deep learning foundations
Prompt engineering and OpenAI API integration
NLP with SpaCy: entity recognition, text classification, and pipelines

Support designed to keep you progressing

13-week part-time mentored format — study around your schedule
Video lessons and notebooks to work through at your own pace
Regular 1:1 sessions with your instructor to review your work
Dedicated support channel for questions between sessions
Hands-on challenges at the end of every module
Industry-recognized graduation certificate

Your mentor

Dr. Minerva Singh

Minerva holds a PhD from the University of Cambridge, an MPhil from the School of Geography and Environment, and an MSc from the Department of Engineering at Oxford University.

With more than a decade of academic research and mentoring experience, she brings deep practical knowledge in machine learning, deep learning, NLP, LLMs, and data science with Python and R.

She has contributed to peer-reviewed journals including PLOS One, reviewed for journals such as Remote Sensing, and given guest lectures at events like Open Data Science Conference.

The outcomes

After the bootcamp you will be ready to start applying for junior roles such as:

Machine Learning Engineer
AI Engineer
LLM / Generative AI Engineer
NLP Engineer
Deep Learning Engineer
Data Scientist
Computer Vision Engineer
AI Application Developer

Curriculum breakdown

Module 1: Python, Data Wrangling and Visualisation

Reinforce Python fundamentals and master the core data tools: NumPy, Pandas, Matplotlib, and Seaborn.

By the end of this module you will be able to load, clean, explore, and visualize real datasets — the foundation every ML engineer builds on.

Module 2: Classical Machine Learning

Learn the core machine learning algorithms that every AI engineer should understand: supervised and unsupervised learning, tree-based models, K-means clustering, and model evaluation with Scikit-Learn.

Module 3: Neural Networks

Get to grips with how neural networks actually work — gradient descent, backpropagation, and optimization. Build and train multilayer perceptrons with Keras.

Project: classify handwritten digits with the MNIST dataset.

Module 4: Convolutional and Recurrent Neural Networks

Work with CNNs for computer vision and RNNs for sequential data. Learn transfer learning with ResNet50.

Real project: end-to-end ship detection in satellite imagery — from exploratory data analysis and traditional computer vision through to a custom CNN and ResNet50 transfer learning model.

Module 5: Generative Networks — Autoencoders and VAEs

Understand self-supervised learning and build generative models from scratch. Cover autoencoders, variational autoencoders (VAEs), and denoising networks.

Project: image denoising with autoencoders.

Module 6: LLMs, NLP, and Modern AI Engineering

The deepest module in the course. Start with NLP foundations using SpaCy — entity recognition, pipelines, keyword extraction, and text similarity — then move into the modern LLM stack.

You will work with the OpenAI API, load and run open-source models via Hugging Face (including Mistral), build AI applications with LangChain and LlamaIndex, and implement full RAG pipelines with embeddings and vector retrieval.

The module also covers prompt engineering, LLMOps, model deployment on GCP, quantization, and fine-tuning with SGD — everything needed to move from experimenting with LLMs to deploying them.

Module 7: Final Project

Build a project of your own choosing in the domain that interests you most — computer vision, NLP, LLM applications, or generative models. Your instructor will help you scope it, review your work, and push you to make it something you can show to a hiring manager.

How to join the bootcamp

Choose your starting date and register online with the 900€ prepayment.
Fill out the student form.
Receive the prep materials including our Python video course.
Join our Slack workspace.
Start preparing before the bootcamp begins.
Make the remaining payment due 5 days before the starting date.
On day 1 receive the workflow instructions and main materials to start working on the course.

FAQ

Will I learn to work with Hugging Face and open-source LLMs?

Yes. Module 6 covers Hugging Face in depth — loading models, running inference, working with Mistral, and fine-tuning. You will also work with the OpenAI API, LangChain, and LlamaIndex as part of the same module.

How does this compare to free online courses like fast.ai or Coursera?

Free courses are great for learning concepts. What they cannot give you is someone who reviews your specific code, answers your specific questions, and adjusts the pace and depth to where you are. The 1:1 mentored format is what gets people unstuck and moving forward — and it is what makes the difference between finishing a course and actually being job-ready.

What is the most advanced thing I will build?

By the end of Module 6 you will have built a working RAG pipeline — a system that retrieves relevant documents and uses an LLM to answer questions over them. You will also have fine-tuned an open-source LLM and deployed a model to GCP. Your final project in Module 7 goes further in whichever direction interests you most.

What computer would I need?

Any operating system will do: Mac, Windows, or Linux. Most of the heavy computation runs on Google Colab, so your local machine does not need to be powerful.

What qualification will I get from this course?

After successful completion you will receive an industry-recognized graduation certificate and be prepared for junior ML and AI engineering roles.

Is there a pre-course?

Yes. We send Python training materials before the bootcamp begins. If you are new to Python, our included video course — around 7 hours of lessons and 100 exercises — will get you to the right level before day one.

How does the price compare to other schools?

At 3,950€ for 13 weeks of 1:1 mentored learning from a Cambridge-trained researcher, this is significantly below what comparable programmes cost elsewhere. Most 1:1 mentored AI bootcamps charge €8,000–€15,000. The difference is our focus on teaching quality over marketing spend.